1 | /* SPDX-License-Identifier: GPL-2.0 */ |
2 | #ifndef _LINUX_NLS_H |
3 | #define _LINUX_NLS_H |
4 | |
5 | #include <linux/init.h> |
6 | |
7 | /* Unicode has changed over the years. Unicode code points no longer |
8 | * fit into 16 bits; as of Unicode 5 valid code points range from 0 |
9 | * to 0x10ffff (17 planes, where each plane holds 65536 code points). |
10 | * |
11 | * The original decision to represent Unicode characters as 16-bit |
12 | * wchar_t values is now outdated. But plane 0 still includes the |
13 | * most commonly used characters, so we will retain it. The newer |
14 | * 32-bit unicode_t type can be used when it is necessary to |
15 | * represent the full Unicode character set. |
16 | */ |
17 | |
18 | /* Plane-0 Unicode character */ |
19 | typedef u16 wchar_t; |
20 | #define MAX_WCHAR_T 0xffff |
21 | |
22 | /* Arbitrary Unicode character */ |
23 | typedef u32 unicode_t; |
24 | |
25 | struct nls_table { |
26 | const char *charset; |
27 | const char *alias; |
28 | int (*uni2char) (wchar_t uni, unsigned char *out, int boundlen); |
29 | int (*char2uni) (const unsigned char *rawstring, int boundlen, |
30 | wchar_t *uni); |
31 | const unsigned char *charset2lower; |
32 | const unsigned char *charset2upper; |
33 | struct module *owner; |
34 | struct nls_table *next; |
35 | }; |
36 | |
37 | /* this value hold the maximum octet of charset */ |
38 | #define NLS_MAX_CHARSET_SIZE 6 /* for UTF-8 */ |
39 | |
40 | /* Byte order for UTF-16 strings */ |
41 | enum utf16_endian { |
42 | UTF16_HOST_ENDIAN, |
43 | UTF16_LITTLE_ENDIAN, |
44 | UTF16_BIG_ENDIAN |
45 | }; |
46 | |
47 | /* nls_base.c */ |
48 | extern int __register_nls(struct nls_table *, struct module *); |
49 | extern int unregister_nls(struct nls_table *); |
50 | extern struct nls_table *load_nls(const char *charset); |
51 | extern void unload_nls(struct nls_table *); |
52 | extern struct nls_table *load_nls_default(void); |
53 | #define register_nls(nls) __register_nls((nls), THIS_MODULE) |
54 | |
55 | extern int utf8_to_utf32(const u8 *s, int len, unicode_t *pu); |
56 | extern int utf32_to_utf8(unicode_t u, u8 *s, int maxlen); |
57 | extern int utf8s_to_utf16s(const u8 *s, int len, |
58 | enum utf16_endian endian, wchar_t *pwcs, int maxlen); |
59 | extern int utf16s_to_utf8s(const wchar_t *pwcs, int len, |
60 | enum utf16_endian endian, u8 *s, int maxlen); |
61 | |
62 | static inline unsigned char nls_tolower(struct nls_table *t, unsigned char c) |
63 | { |
64 | unsigned char nc = t->charset2lower[c]; |
65 | |
66 | return nc ? nc : c; |
67 | } |
68 | |
69 | static inline unsigned char nls_toupper(struct nls_table *t, unsigned char c) |
70 | { |
71 | unsigned char nc = t->charset2upper[c]; |
72 | |
73 | return nc ? nc : c; |
74 | } |
75 | |
76 | static inline int nls_strnicmp(struct nls_table *t, const unsigned char *s1, |
77 | const unsigned char *s2, int len) |
78 | { |
79 | while (len--) { |
80 | if (nls_tolower(t, c: *s1++) != nls_tolower(t, c: *s2++)) |
81 | return 1; |
82 | } |
83 | |
84 | return 0; |
85 | } |
86 | |
87 | /* |
88 | * nls_nullsize - return length of null character for codepage |
89 | * @codepage - codepage for which to return length of NULL terminator |
90 | * |
91 | * Since we can't guarantee that the null terminator will be a particular |
92 | * length, we have to check against the codepage. If there's a problem |
93 | * determining it, assume a single-byte NULL terminator. |
94 | */ |
95 | static inline int |
96 | nls_nullsize(const struct nls_table *codepage) |
97 | { |
98 | int charlen; |
99 | char tmp[NLS_MAX_CHARSET_SIZE]; |
100 | |
101 | charlen = codepage->uni2char(0, tmp, NLS_MAX_CHARSET_SIZE); |
102 | |
103 | return charlen > 0 ? charlen : 1; |
104 | } |
105 | |
106 | #define MODULE_ALIAS_NLS(name) MODULE_ALIAS("nls_" __stringify(name)) |
107 | |
108 | #endif /* _LINUX_NLS_H */ |
109 | |
110 | |